Radio pulsars are often used as clocks in a wide variety of experiments. Imperfections in the clock, known as timing noise, have the potential to reduce the significance of, or even thwart e.g. the attempt to find a stochastic gravitational wave (GW) background. We measure the timing noise in a group of 129 mostly middle-aged pulsars (i.e. characterstic ages near 1 Myr) observed with the Parkes radio telescope on a monthly basis since 2014. We examine four different metrics for timing noise, but it remains unclear which, if any, provides the best determination. In spite of this, it is evident that these pulsars have significantly less timing noise than their younger counterparts, but significantly more than the (much older) millisecond pulsars (MSPs). As with previous authors, we find a strong correlation between timing noise and the pulsar spin-down rate, ν. However, for a given ν there is a spread of about a factor 30 in the strength of the timing noise likely indicating that nuclear conditions in the interior of the stars differs between objects. We briefly comment on the implications for GW detection through pulsar timing arrays as the level of timing noise in MSPs may be less than predicted.

Pulsars are small, dense stars which rotate up to 1500 times per second and emit radio waves in a directed beam along their magnetic axis. Pulsars are remarkably stable rotators, and by measuring the arrival times of the radio beam the pulsar can be used as a clock in space. Applications of pulsar timing have led to the first exoplanet system, stringent tests of theories of gravity and have the potential to detect gravitational waves. Yet, the pulsars spin is not perfect, and understanding the imperfections (timing noise) is important to the aforementioned applications. Here, we present the analyses of timing noise of 133 pulsars observed with the Parkes radio telescope in Australia over the period of 4 years. The results show that as consistent with pulsar toy model equations the measured breaking index, representing the magnitude of timing noise, has an anti-correlation with the spin-down rate, spin-down energy and characteristics magnetic fields, and it also has a positive correlation with the characteristics age at significant of 10σ. The error in the slope values of those relationships are in the order of 9 percent, which may imply imperfections of the pulsar toy model.